Use of Clinical Samples to Identify Influenza Virus Antigenic Variants Summary Influenza A viruses (IAVs) cause pandemic and seasonal outbreaks that lead to the loss of thousands to millions of human lives. Vaccination is the best option for preventing influenza outbreaks and minimizing their effects. An understanding of the antigenic evolution of influenza viruses and the rapid selection of a well- matched influenza vaccine strain is the key to developing an effective vaccination program. However, antigenic characterization for influenza viruses presents two great challenges: 1) virus propagation, which is required in conventional serologic assays, can cause culture-adapted mutations and skew antigenic properties of viruses in clinical samples, and 2) reference sera used in conventional serologic assays are produced in influenza virus?seronegative ferrets and do not represent the immunologic profiles of human serum, which often has had prior exposures to influenza viruses through vaccination, natural infection, or both. An ideal platform for determining antigenic properties of influenza viruses and for selecting influenza vaccine strain should directly use clinical samples. The objectives of this project are 1) to develop and apply a novel high-throughput technology to directly characterize antigenic properties of influenza viruses by using human clinical samples without virus isolation and propagation and 2) to understand antigenic evolution of IAVs by using clinical samples directly. The antigenic characterization will include influenza virus?positive clinical samples from which virus can or cannot be cultivated. To understand influenza virus quasispecies in clinical samples and the effect of culture-adapted mutations on antigenic characterization, we will perform next-generation genomic sequencing on the clinical samples and corresponding isolates. We will then study the effects of the sequence diversity on antigenic variations of influenza viruses. We will also determine the effect that prior exposure to influenza virus(es) has on antigenic characterization during influenza vaccine strain selection. This project will help us provide fundamental technology for characterizing the antigenicity of influenza viruses in clinical samples without propagating virus. The resulting platform for antigenic characterization will overcome biases arising from virus propagation in conventional serologic assays. In addition, this is a high- throughput method and will significantly reduce the human labor needed for serologic characterization, decrease the time required for antigenic characterization, and increase the number of samples in antigenic characterization. Thus, this project will lead to significant technologic advances in influenza vaccine strain selection and facilitate influenza prevention and control. In addition, this project will provide knowledge about molecular mechanisms in antigenic variations associated with influenza virus quasispecies and genomic diversity and knowledge about prior human exposure to influenza viruses, which will help us optimize antigenic characterization in vaccine strain selection and understand antigenic evolution of influenza viruses.